Salmonella enterica gehört zu den gramnegativen Bakterien und ist der Erreger von verschiedenen Darmerkrankungen, von Gastroenteritis bis systemische Infektionen, und jährlich die Ursache für hunderttausende Todesfälle. In den letzten 30 Jahren wurden grundlegende Mechanismen der Invasion und des intrazellulären Wachstums von Salmonella gelöst: Salmonella verfügt über eine Reihe von Virulenzfaktoren (Effektorproteine), die mittels Typ-III-Sekretionssystemen (T3SS) zur molekularen Manipulation der Wirtszelle ausgeschüttet werden. Allerdings sind die Funktionen einiger dieser Effektorproteine nur unzureichend charakterisiert. Darüber hinaus hat die Rolle von RNA-Protein-Wechselwirkungen in bakteriellen Prozessen, einschließlich Infektionen, an Bedeutung gewonnen. Eine der am häufigsten verwendeten Techniken zur Untersuchung von Effektorproteinen ist der Gentamicin Protection Assay (GPA), eine einfache Methode, die den Salmonella-Infektionsprozess in vitro nachbildet. Da der GPA jedoch starken Schwankungen unterliegt und eine Endpunktmessungen darstellt, ist dieser unzureichend, wenn gleichzeitig mehrere Bedingungen oder zeitabhängige Dynamiken untersucht werden müssen. Um diese Einschränkungen zu umgeben, wurde InFluence entwickelt, eine Hochdurchsatz-Fluoreszenz-Mikroskopiemethode. Diese Methode ermöglicht Einblicke in die von Salmonellen besiedelten intrazellulären Nischen, und erwies sich als nützliches Instrument zur Charakterisierung von nicht nur von Effektorproteinen, sondern auch von Wirtsproteinen im Infektionsprozess. Darüber hinaus haben wir zur Entwicklung von TriPepSVM, einer Support Vector Machine (SVM), beigetragen. Dieser Algorithmus, der zusammen mit der AG Marsico (ICB - Helmholtz Zentrum München) entwickelt wurde, sagt RNA-Bindeproteine voraus, mithilfe von Tripeptiden in intrinsisch ungefalteten Regionen (IDR). 66 RBPs hat TriPepSVM in Salmonella vorausgesagt, wovon drei im Rahmen dieser Arbeit experimentell validiert wurden. / Salmonella enterica is a species of Gram-negative bacteria and the causative agent of enteric diseases, ranging from gastroenteritis to systemic infections, causing hundreds of thousands of deaths every year. In the last 30 years the basic mechanisms underpinning invasion and intracellular growth have been unravelled: Salmonella produces tens of virulence factors - termed “effectors” - which are secreted by two distinct Type III Secretion Systems (T3SS) for the hijacking of the host cell molecular machinery via protein-protein interactions. Although the biochemical activities of many effectors have been characterised, the functions of some have remained elusive. In addition, post-transcriptional regulation has emerged to prominence in bacteria, with RNA-protein interactions playing a pivotal role in many bacterial functions, including infection. One of the most frequently used techniques to study Salmonella effectors is the Gentamicin Protection Assay (GPA), a simple method which replicates the infection process in vitro; however, as GPA is subject to high levels of variation and only generates end-point measurements, the method can be inadequate when studying multiple conditions at once or following time-dependent dynamics. InFluence, a high-throughput, fluorescence microscopy-based analysis pipeline, was designed to address these issues. InFluence offered insights into the intracellular niches occupied by Salmonella, and proved to be a useful tool in assessing not only the contribution to the infection process of effectors, but that of host proteins too. In addition, we contributed to the development of TriPepSVM, a support vector machine-based (SVM) method developed with the Marsico group (Institute of Computational Biology (ICB) - Helmholtz Zentrum Munchen) for predicting RNA-binding proteins (RBPs) using tripeptides that are frequent in Intrinsic Disordered Regions (IDRs). TriPepSVM predicted 66 new RBPs in Salmonella, and three were experimentally validated.

<p>  </p> <p>Hyperspectral Imaging is one of the most popular technologies in plant phenotyping due to its ability to predict the plant physiological features such as yield biomass, leaf moisture, and nitrogen content accurately, non-destructively, and efficiently. Various kinds of hyperspectral imaging systems have been developed in the past years for both greenhouse and field phenotyping activities. Developing the plant physiological prediction model such as relative water content (RWC) using hyperspectral imaging data requires the adoption of machine learning-based calibration techniques. Convolutional neural networks (CNNs) have been known to automatically extract the features from the raw data which can lead to highly accurate physiological prediction models. Once a reliable prediction model has been developed, sharing that model across multiple hyperspectral imaging systems is very desirable since collecting the large number of ground truth labels for predictive model development is expensive and tedious. However, there are always significant differences in imaging sensors, imaging, and environmental conditions between different hyperspectral imaging facilities, which makes it difficult to share plant features prediction models. Calibration transfer between the imaging systems is critically important. In this thesis, two approaches were taken to address the calibration transfer from the greenhouse to the field. First, direct standardization (DS), piecewise direct standardization (PDS), double window piecewise direct standardization (DPDS) and spectral space transfer (SST) were used for standardizing the spectral reflectance to minimize the artifacts and spectral differences between different greenhouse imaging systems. A linear transformation matrix estimated using SST based on a small set of plant samples imaged in two facilities reduced the root mean square error (RMSE) for maize physiological feature prediction significantly, i.e., from 10.64% to 2.42% for RWC and from 1.84% to 0.11% for nitrogen content. Second, common latent space features between two greenhouses or a greenhouse and field imaging system were extracted in an unsupervised fashion. Two different models based on deep adversarial domain adaptation are trained, evaluated, and tested. In contrast to linear standardization approaches developed using the same plant samples imaged in two greenhouse facilities, the domain adaptation extracted non-linear features common between spectra of different imaging systems. Results showed that transferred RWC models reduced the RMSE by up to 45.9% for the greenhouse calibration transfer and 12.4% for a greenhouse to field transfer. The plot scale evaluation of the transferred RWC model showed no significant difference between the measurements and predictions. The methods developed and reported in this study can be used to recover the performance plummeted due to the spectral differences caused by the new phenotyping system and to share the knowledge among plant phenotyping researchers and scientists.</p>